System and methods for a measurement tool for medical imaging

Abstract

Methods and systems are provided for evaluating a subject for a liver disease using ultrasound images. In one example, a method includes, in response to a request to evaluate the liver disease, determining, with a measurement model, that a selected medical image frame of the subject includes a target anatomical view and has an image quality above a threshold image quality, and in response, measuring, with the measurement model, a marker for the liver disease in the selected medical image, and outputting, for display on a display device, the measurement of the marker.

1 . A method for evaluating a liver disease of a subject, comprising:
 in response to a request to evaluate the liver disease, determining, with a measurement model, that a selected medical image frame of the subject includes a target anatomical view and has an image quality above an image quality threshold, and in response, measuring, with the measurement model, a marker for the liver disease in the selected medical image frame; and   outputting, for display on a display device, a measurement of the marker,
 wherein the marker is visceral fat surrounding a liver, and 
 wherein the measurement model is trained to measure a thickness of the visceral fat. 
   
     
     
         2 . The method of  claim 1 , further comprising determining, with the measurement model, that a prior medical image frame of the subject acquired before the selected medical image frame does not include the target anatomical view and/or does not have an image quality above the image quality threshold, and in response, rejecting the prior medical image frame such that the marker is not measured. 
     
     
         3 . The method of  claim 2 , wherein the measurement model comprises one or more deep learning networks and wherein determining that the selected medical image frame includes the target anatomical view and has an image quality above the image quality threshold comprises entering the selected medical image frame into one or more of the one or more deep learning networks, where the one or more deep learning networks are trained to output an anatomical view of the selected medical image frame and an image quality of the selected medical image frame. 
     
     
         4 . The method of  claim 3 , wherein measuring the marker of the liver disease comprises entering the selected medical image frame into the one or more deep learning networks, where the one or more deep learning networks are trained to segment the selected medical image frame in order to identify the marker and measure the identified marker. 
     
     
         5 . The method of  claim 3 , wherein the image quality threshold is learned by the one or more deep learning networks during training. 
     
     
         6 . The method of  claim 5 , wherein to learn the image quality threshold and to output the image quality, the one or more deep learning networks are trained with a training dataset that includes a plurality of sets of medical image frames, each including a high quality image frame and a subset of image frames generated from the high quality image frame, each at a progressively lower quality. 
     
     
         7 . The method of  claim 1 , wherein the liver disease is non-alcoholic fatty liver disease. 
     
     
         8 . The method of  claim 1 , further comprising determining a risk that the subject has the liver disease or a current progression of the liver disease in the subject based on the measurement of the marker, and outputting, for display, a visual representation of the risk or the progression. 
     
     
         9 . A method for evaluating a liver disease of a subject, comprising:
 in response to a request to evaluate the liver disease, determining, with a measurement model, that a selected ultrasound image of the subject includes a target anatomical view and has an image quality that is sufficient for performing a measurement of visceral fat in the selected ultrasound image, and in response, measuring, with the measurement model, a thickness of the visceral fat in the selected ultrasound image, where the measurement model is trained to determine that the image quality is sufficient for performing the measurement of the visceral fat using a training dataset that comprises a plurality of subsets of images, each subset of images comprising an initial image and one or more reduced quality images generated from the initial image; and   outputting, for display on a display device, the measurement of the thickness of the visceral fat.   
     
     
         10 . The method of  claim 9 , wherein the measurement model is trained to output an image quality metric indicating the image quality of the selected ultrasound image and determine that the image quality is sufficient based on the image quality metric being greater than a sufficiency threshold, wherein the sufficiency threshold is learned by the measurement model using the training dataset. 
     
     
         11 . The method of  claim 9 , further comprising determining a risk that the subject has the liver disease or a current progression of the liver disease in the subject based on the measurement, and outputting, for display, a visual representation of the risk or the progression. 
     
     
         12 . The method of  claim 9 , wherein the liver disease is non-alcoholic fatty liver disease. 
     
     
         13 . The method of  claim 9 , wherein the measurement model comprises one or more deep learning networks. 
     
     
         14 . The method of  claim 9 , wherein measuring, with the measurement model, the thickness of the visceral fat in the selected ultrasound image comprises entering the selected ultrasound image as input to the measurement model, the measurement model trained to segment the selected ultrasound image to identify the visceral fat, place measurement points on borders of the visceral fat, and determine the thickness by measuring a distance between the measurement points.